Prompt Engineering Is Creating a New Enterprise AI Attack Surface

Mayank Kumar, founding AI engineer at DeepTempo, explains how the rise of prompt engineering is introducing a new class of security risks in enterprise IT environments. As organizations operationalize generative AI, poorly governed prompts, model manipulation and indirect injection attacks are expanding the threat surface, requiring new controls, observability and policy frameworks.